Perl 6 language and compiler development | Logs at | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
01:13 anatofuz left 01:25 leont left, anatofuz joined 01:26 anatofuz left, anatofuz joined 01:43 lucasb left 02:00 anatofuz_ joined 02:04 anatofuz left 02:22 anatofuz_ left 02:23 anatofuz joined, anatofuz left 02:28 anatofuz joined, anatofuz left 02:43 anatofuz joined, anatofuz left 02:47 anatofuz joined, anatofuz left 02:50 anatofuz joined 02:52 anatofuz left 02:57 anatofuz joined, anatofuz left 02:59 anatofuz joined, anatofuz left 03:33 anatofuz joined 03:34 anatofuz left 03:35 anatofuz joined 04:54 anatofuz left 04:55 anatofuz joined 04:59 anatofuz left 05:03 robertle left 05:16 anatofuz joined 06:26 Geth joined, Geth_ left
moritz it might have been mentioned before, but might be interesting for those working on schedulers :D 06:27
06:30 anatofuz left 06:42 anatofuz joined 06:46 anatofuz left 06:51 anatofuz joined 06:52 anatofuz left 06:53 anatofuz joined 07:22 anatofuz_ joined, anatofuz_ left 07:23 anatofuz_ joined 07:25 anatofuz left 07:52 anatofuz joined 07:54 anatofuz_ left 08:45 anatofuz left 08:47 anatofuz joined 08:49 anatofuz left 08:50 anatofuz joined 09:15 squashable6 left 09:17 squashable6 joined, ChanServ sets mode: +v squashable6 10:29 epony left 11:23 epony joined
|Tux| Rakudo version 2019.07.1-398-gf3dda96a8 - MoarVM version 2019.07.1-268-g07b00bbd7
csv-ip5xs0.762 - 0.779
csv-ip5xs-206.408 - 6.425
csv-parser21.066 - 21.125
csv-test-xs-200.422 - 0.428
test6.942 - 7.102
test-t1.743 - 1.746
test-t --race0.778 - 0.808
test-t-2029.182 - 30.079
test-t-20 --race9.075 - 9.183
Is there anything I need to change in Text::CSV to support raku?
lizmat are you using .perl ? 11:42
in the further future, the name of the executable that you're calling
other than that, I don't think anything is being planned on being changed 11:43
off the top of my hat
11:56 Kaiepi left 12:23 lucasb joined 12:25 Kaiepi joined
AlexDaniel |Tux|: maybe post those timing on #raku-dev from now on :) 12:35
|Tux|: also if you have mentions of Perl 6 in the readme and elsewhere, you should at least mention Raku if not just Raku 12:36
12:42 |Tux| joined 13:30 squashable6 left 13:31 squashable6 joined, ChanServ sets mode: +v squashable6 13:35 epony left 13:42 epony joined
[Coke] regarding the raku* channels, is someone like mst on board with permissions or registration or whatever IRC needs? 14:09
lizmat AlexDaniel: ^^ 14:15
AlexDaniel [Coke]: yes
[Coke]: 14:16
[Coke] +1 14:19
14:34 MasterDuke joined
releasable6 Next release in ≈4 days and ≈3 hours. R6 is down. At least 5 blockers. Please log your changes in the ChangeLog: 15:00
15:00 anatofuz_ joined 15:03 anatofuz left 15:05 anatofuz_ left
japhb What is the "R6 is down" message about? 15:50
16:01 robertle joined
AlexDaniel japhb: it's about 16:02
japhb: it's no longer relevant because RT tickets will be transferred to github
16:20 releasable6 joined, ChanServ sets mode: +v releasable6
lizmat and the last Perl 6 Weekly hits the Net: 16:48
sjn yay! 16:51
rba What was on ? 16:59
AlexDaniel rba: perl6 tickets, same as on 17:26
rba: it was just a different interface
rba: it was a single page app, pretty much, so if you're worried about broken links to it you can maybe redirect it somewhere 17:27
in the future it could lead to
but not now
18:59 squashable6 left 19:02 squashable6 joined 19:57 hankache joined 20:01 hankache left 20:04 hankache joined 20:35 MasterDuke left 20:58 hankache left 21:22 patrickb joined, anatofuz joined
timotimo we may want to think about FALLBACK contributing via roles, not just TWEAK and BUILD 21:24
21:41 squashable6 left 21:43 squashable6 joined, ChanServ sets mode: +v squashable6
patrickb Currently the perl6 executable that is run from within the build dir creates a folder structure `../share/perl6/[core,site,vendor]/short`relative to the build directory. This is obviously unwanted. 21:50
I'm pretty sure `src/core.c/CompUnit/RepositoryRegistry setup-repositories()` is to blame. 21:51
I'm not sure what a good fix might be though. We could either detect when perl6 is run from the build dir and special case on that, that sounds unclean to me. Or we hit at perl6 that the repository is located in a folder where it does little harm (e.g. the gen/ folder). 21:54
Any opinions?
I think this bug has been there for a long time, but has previously created that folder structure in the installation path. 21:56
vrurg timotimo: if FALLBACK is a method it'd be copied down to consuming class anyway. 21:57
patrickb: how do you want to hint it?
patrickb It's basically using PERL6_HOME to create the folder structure. A environment variable (could be passed in the perl6-m.c build runner) is enough to point it at a different folder. 21:59
The reason it puts it in ../share/perl6/ is because it tries to determine PERL6_HOME relative to the executable path. But that is wrong when the executable is not located in an installation. 22:00
vrurg patrickb: perhaps, binaries must be created in a subdir? But I foresee some strong opposition to this. 22:02
Though the right thing would be to make blib the build destination for everything. Binaries would the go into blib/bin 22:03
patrickb As long as we don't replicate an installation folder structure inside the build dir we do have to put hacks in to make it work.
vrurg So, we probably must replicate it. 22:04
patrickb This is gonna get lots of opposition. 22:05
vrurg timotimo: I think if we make submethod FALLBACK work in roles there must be a set of strict rules on how mutlple fallbacks are handled. Basically, it would have to be another multidispatch. 22:06
timotimo yeah 22:07
vrurg patrickb: Yes, no doubt.
patrickb Going all the way would mean not only putting the perl6 executable somewhere else, but all the native and p6 libraries just as well.
vrurg timotimo: so, basically, it could be a special case where we implicitly create a proto and proclaim that FALLBACK must always be multi. Then we can just collect them from class roles and send into the proto as candidates. 22:08
patrickb: I think it needs a well made plan and a problem-solving ticket. 22:09
patrickb Do you think it makes sense to add a quick patch that makes it generate the dummy folders in build-dir/gen/ ? 22:10
timotimo a wall-made plan would potentially also do
patrickb Until we arrive at a larger solution?
vrurg patrickb: If there are no urgent complants, I'd leave it as-is for a while. Because there is nothing more permanent than a temporary... 22:11
patrickb Oh! You have that saying in the US as well? 22:12
vrurg patrickb: We have this saying in eastern slavic langs. ;) But it's likely coming from Roman times, I guess.
22:12 Kaiepi left 22:13 Kaiepi joined
patrickb Well, it's placing folders in a random directory of the user. I suspect most people have just wondered where the folders came from, but didn't make the connection. 22:13
AlexDaniel timotimo, patrickb, vrurg: any change we can have these great conversations on #raku-dev? :) 22:15
chance* 22:16
vrurg Oops. I thought we're there. :D
patrickb: let's move on, reallly.
22:18 Kaiepi left 22:19 Kaiepi joined
vrurg timotimo: would join us on #raku-dev? I'd like to ask you a few questions, if you don't mind. 22:19
AlexDaniel also please configure your irc clients to autojoin #raku and #raku-dev and also perhaps remove #perl6 and #perl6-dev from the list :) 22:21
22:23 Kaiepi left
vrurg I don't see a reason of not autojoining #perl6* until we declare it deprecated and setup a bot to redirect people to the right place in one or another way. 22:24
AlexDaniel vrurg: it's deprecated, leave :P 22:26
vrurg Where is the bot kicking my ass into the right direction? Hedgehog is a lazy bird, not fliying without a kick! :D 22:27
AlexDaniel vrurg: beep boop #raku beep boop
vrurg AlexDaniel: zzzzz... beep-beep-beep... zzzzzzz... :D :D :D 22:28
22:59 patrickb left